Virgil — AI skill for Claude Code
Runtime de controle open source para agentes de IA: um Edge Gateway local intercepta chamadas a provedores LLM (OpenAI, Anthropic, Kimi e compatíveis), aplica políticas determinísticas de custo/segura.
How to install Virgil
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open MiguelReis944/Virgil and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Virgil does
Runtime de controle open source para agentes de IA: um Edge Gateway local intercepta chamadas a provedores LLM (OpenAI, Anthropic, Kimi e compatíveis), aplica políticas determinísticas de custo/segurança, registra telemetria com redaction em SQLite e opcionalmente exporta métricas a um Control Plane. Local-first, sem conta ou Docker obrigatórios.
Alternatives in AI
- Axonhub — ⚡️ Open-source AI Gateway — Use any SDK to call 100+ LLMs 3.2k ★
- WindsurfAPI — Turn Windsurf / Devin Desktop's 100+ AI models (Claude, GPT, Gemini, DeepSeek, Kimi, GLM, SWE) into OpenAI-, A 3k ★
- Vllm Mlx — High-performance OpenAI and Anthropic compatible LLM inference server for Apple Silicon 1.5k ★
README
Virgil
Virgil is a local circuit breaker for supervised AI agents. One executable runs the core, gateway, and browser panel on the same machine. A policy block is recorded and terminates the supervised process tree. The panel is available at `/dashboard` on the loopback address.
Current functionality
The gateway supports configured OpenAI-compatible requests, provider adapters, local policy checks, SQLite event and execution history, and JSONL export. The core binds to the configured local address. Setup is served locally when the configuration is missing.
`virgil` starts the local core and opens the browser panel; `virgil serve` remains an alias. `virgil run` starts a child under core supervision, supplies the gateway address and run credential, and terminates its process tree on a policy block, deadline, or interruption. `virgil version` displays the binary's version, commit, and build date.
The panel includes Providers, Protections, Executions, Health, Settings, and Usage. Start with Providers and Protections, then launch a supervised child and inspect its result under Executions.
This local product does not require an account or hosted Control Plane. It currently binds to loopback; secure LAN operation is a separate future stage. Historical team-service concepts are retained later in this README as deferred design context.
For ready-to-run binaries, checksums, setup, backup, and upgrade instructions, see [Local installation](docs/install.md). The [release acceptance checklist](docs/release-acceptance.md) records the tests required before publishing a release.
Build and start the local core
Install Go 1.26 or newer. From the repository, build and start Virgil with:
go build -o virgil ./cmd/virgil
./virgil --config configs/virgil.example.toml
On Windows, run the resulting `virgil.exe`. For development without building a binary, use `go run ./cmd/virgil --config configs/virgil.example.toml`. The example configuration uses a local addres
Related Skills
Devo
Model-neutral agent desktop/runtime for private, enterprise, and OpenAI-compatible / Anthropic-compatible mode
Agent Network
助力搭建你的数字 AI 员工军团 — 多 Agent 一行命令组网协作。Claude Code / Claude Agent SDK / Codex / Grok Build 4 runtime + 8+ 家 LLM(A
Cab
One local gateway for every coding agent. Cost-aware LLM model routing for Claude Code, Codex, OpenCode, and m
Open Managed Agents
Self-hostable, open-source alternative to Claude Managed Agents. Multi-LLM support (Anthropic, OpenAI, Ollama)
Agent Rs
Pure-Rust async runtime for LLM agents. Multi-provider (Anthropic/OpenAI/Ollama), tool-capable end-to-end, str
Token Proxy
Local AI API gateway for OpenAI / Gemini / Anthropic. Runs on your machine, keeps tokens counted (SQLite), off
Related Agents
Langchain Prompt Designer
Especialista en system prompts de agentes LangChain (LLM nodes en N8N, OpenAI Agents, Anthropic, etc.). Diseña
AI LLM Engineer
Use this agent when the user types /ai or asks for ai/llm engineer work — e.g., build a RAG pipeline over our
LLM Orchestrator
Use this agent for LLM integration work — prompt engineering, multi-provider abstraction (OpenRouter, Gemini,